OUP user menu

A comparison of two invasive strategies in patients with non-ST elevation acute coronary syndromes: results of the Early or Late Intervention in unStable Angina (ELISA) pilot study
2b/3a upstream therapy and acute coronary syndromes

Arnoud W. J. van ’t Hof, Suzanna T. de Vries, Jan-Henk E. Dambrink, Kor Miedema, Harry Suryapranata, Jan C. A. Hoorntje, A.T. Marcel Gosselink, Felix Zijlstra, Menko-Jan de Boer
DOI: http://dx.doi.org/10.1016/S0195-668X(03)00259-8 1401-1405 First published online: 1 August 2003


Background Only few studies specifically addressed the effect of timing of angiography and/or pre-treatment with a glycoprotein 2b/3a receptor blocker in patients with non-ST elevation acute coronary syndromes (ACS) who undergo invasive treatment.

Methods In a 2-year period, 220 patients with non-ST elevation ACS, were randomized to early angiography without tirofiban pre-treatment (Early strategy) or to delayed angiography after 24–48h pre-treatment with tirofiban (Late strategy). The first 48h after admission, CKmb levels were measured and enzymatic infarct size (LDHQ48) was assessed by the area under the LDH release curve. When PCI was performed beyond 48h, CKmb was measured 6 and 12h afterwards.

Results Median time to angiography was 6 (Early) and 50 (Late) hours. PCI was performed in 130 patients (59%). In these patients, a patent (TIMI 2 or 3 flow) culprit vessel was more often present in the Late group compared to the Early group (66% vs 82% p=0.05). In patients with an elevated CKmb (n=96, 44%), LDHQ48was significantly lower in patients who underwent angiography after pre-treatment with tirofiban (629±503U/L (Early) vs 432±441U/L (Late), p=0.02). No difference in clinical outcome between the groups was observed at 30 days follow-up.

Conclusion This pilot study showed that a strategy of delayed angiography with concomitant pre-treatment with tirofiban is associated with improved angiographic outcomes and less initial enzyme release, compared to a strategy of immediate angiography without 2b/3a inhibitor pre-treatment. The use of an end point parameter, which assess total enzyme release over a given period of time, might be of special value in patients with non-ST elevation ACS, who undergo very early invasive treatment.

  • Non ST elevation ACS
  • Anti-platelet therapy
  • Enzymatic infarct size

1 Introduction

Recent studies show that an aggressive strategy of early angiography and revascularisation may improve outcome in high-risk non-ST elevation acute coronary syndrome (ACS) patients.1,2However, in the FRISC trial, earlyintervention was associated with an increased incidenceof myocardial infarction in the first days, compared to a conservative management. This was consistent with the findings from the older angioplasty trials, which showed that early intervention in patients with unstable angina was associated with a high incidence of adverse events.3Patients with unstable angina often have (angiographically or angioscopically) evidence of thrombi and it is known that these thrombi increase the risk of percutaneous coronary intervention (PCI).4Treatment with thrombolysis, however, did not improve results.5The TACTICS TIMI 18 study showed that the hazard of myocardial infarction was less obvious when patients were pre-treated with the glycoprotein 2b/3a receptor blocker tirofiban during 24h.2Only few studies specifically addressed the effect of timing of angiography and of prolonged pre-treatment with a glycoprotein 2b/3a blocker (‘upstream’ therapy) before coronaryangiography. Very early intervention might reduce myocardial damage in case of an occlusive culprit lesion, and might prevent early recurrent ischaemic events, whereas delayed angiography after pre-treatment therapy may lower the risk of intervention after a cool-down period. This study compares two strategies in patients with unstable angina: immediate intervention without pre-treatment with tirofiban versus delayed intervention after prolonged pre-treatment withtirofiban.

2 Patients and methods

Inclusion criteria were the presence of at least two of the following criteria: unstable angina pectoris, not induced by a non-cardiac condition, with the last episode of pain within the preceding 24h, ST segment depression >1mm or a positive troponin T (>0.05ng/ml) on admission or 3h later. Patients with negative T waves only, patients over 80 years of age, patients who had a PCI within the preceding 6 months and patients with renal failure, a contra-indication to treatment with a 2b/3a inhibitor or patients in Killip class 3 or 4 were excluded. Patients randomized to early intervention were scheduled for angiography within 12h without tirofiban pre-treatment (Early group). The other group was pre-treated with tirofiban (Merck, West Point, PA.). Until 31 December 2000, a loading dose of 0.4μg/kg/30min, followed by a maintenance infusion of 0.1μg/kg/min was given for 24h. After the results of the TARGET study,6it was decided to continue the study with a higher dosage of 10μg/kg loading bolus followed by 0.15μg/kg/min maintenance infusion until angiography, with continuation of tirofiban for at least 12h in case PCI was performed (Late group). Both groups were treated with Aspirin (500mg iv.), subcutaneous low-molecular weight heparin, oral β-blockers and nitroglycerin intravenously. After stenting all patients received 75mg of clopidogrel daily after a loading dose of 300mg. Treatment with low molecular weight heparin was given for at least 48h and was continued for another 48h in case PCI was performed. All angiographic data were analysed by a core-lab, unaware of treatment modality (Diagram, Zwolle, The Netherlands). TIMI flow and TIMI frame count were assessed only in those patients who underwent PCI, and these parameters were evaluated as described by Gibsonet al.7

3 Enzyme release and enzymatic infarct size

Blood samples were obtained on admission and every 6–12h hereafter up to 48h. In case PCI was performed later than 48h after admission, cardiac enzymes were measured at 6 and 12h after PCI. Enzymatic infarct size (LDHQ48) was calculated in those patients with a positive Creatin Kinase MB fraction (CKmb) within the first 48h of admission. This enzymatic infarct size is based on enzyme concentrations of Lactate Dehydrogenase (LDH) as the reference enzyme, in which an area under the curve is calculated from preferably seven, but from at least five measurements. A two-compartment model was used, which has been validated in studies on the turnover of radiolabelled plasma proteins and circulating enzymes.8–10

4 Clinical outcome

Total death and recurrent myocardial infarction were assessed at 30 days follow-up. Spontaneous recurrent myocardial infarction was defined as a new CKmb rise of more than 6% of total CK, whenever CK was above 200U/l (men) or above 170U/l (women). After PCI, recurrent myocardial infarction was defined as a new rise in CKmb of more than three times the above mentioned 6% level. After CABG, a new myocardial infarction was defined as the development of new Q waves on the post CABG electrocardiogram. Major bleeding was defined as the need for at least 2 Units of blood and a fall in Haemoglobin of more than 2mmol/l, corrective groin surgery, gastro-intestinal or retroperitoneal bleeding.

5 Statistical analysis

5.1 Sample size calculation

The primary end point was enzymatic infarct size. Our hypothesis was that a strategy of very early intervention would reduce enzymatic infarct size by 40%. It was calculated, assuming an infarct size of 500IU/l for the Late and 350IU/l for the Early group and a normal distribution of enzymatic infarct size (this is known from patients with ST elevation ACS8), that 350 patients were needed to prove superiority (to rejectthe null hypothesis: H0 LDHQ48late≠LDHQ4848early−150, H1 LDHQ48late>LDHQ48early−150) of immediate intervention over delayed intervention with a power of 80% and assuming a type I error of 5%.

5.2 Final analysis

The primary end point and other continuous variables were compared using the Student’s t-test and expressed as means±SD, except for time, which was expressed as median and 25 and 75 percentiles. The first secondary end-point, a composite of death and (recurrent) myocardial infarction, is a discrete variable and was compared using Chi square analysis. The Fisher’s exact test was used when the expected cell value was <5. All patients were analysed, including those who never received pre-treatment, or in whom the procedure could not be performed, according to the intention to treat principle.

6 Results

The study was approved by our institutions Review and Ethical Board and all patients gave written informed consent. From April 2000 to December 2001, 220 patients were randomized to early intervention without pre-treatment (Early group) or intervention after 24–48h of pre-treatment with tirofiban (Late group). During this period, 12 patients (5%), who met all in- and exclusion criteria, were not randomized, mainly because of logistic reasons. Baseline characteristics, treatment, and procedural outcome are shown in Table 1. Angiography was performed in almost 100% of patients in both groups. The median time to angiography (25–75 percentile) was 6 (4–14) and 50 (42–73) hours for the Early and Late group respectively. In-hospital revascularization (PCI or CABG) was performed in 75% of Early treated and in 77% of Late treated patients (Table 1). In the 130 patients whounderwent angioplasty after initial angiography, the culprit vessel was patent (TIMI 2 or 3 flow) in 66% (43/66) of patients in the Early group and in 82% (53/64) of patients in the Late group (p=0.05). The procedure was successful (TIMI 3 flow) in 91% of patients in the Early group and in 94% in the Late group patients (p=0.51).The corrected TIMI frame count after PCI of the culprit vessel was23±14 in the Early group and 18±8 in the Late group (p=0.05, Table 1, angiographic outcome). In nine patients in the Early group (8%), tirofiban was given as rescue treatment, because of thrombus, no-reflow or distal embolization.

View this table:
Table 1

Baseline characteristics, treatment and angiographic outcome

Early GroupLate Groupp
Baseline characteristicsn=109n=111
Mean age (yrs±SD)63±1165±110.12
Male Gender (%)79(72)76(68)0.51
Diabetes (%)16(15)16(14)0.96
Hypertension (%)49(45)43(39)0.32
Smoking (%)40(37)36(33)0.54
Previous Myocardial Infarction (%)19(17)14(13)0.32
Previous CABGa(%)12(11)8(7)0.33
Previous PCIb(%)16(15)16(14)0.96
Symptom onset—Admission (min) (25–75 perc)210(120–360)240(135–405)0.64
Angiogram (%)108(99)111(100)
Time to angiogram (h) (25–75 perc)6(4-14)50(42-73)
PCI (%)66(61)64(58)0.60
CABG (%)15(14)21(19)0.32
Conservative (%)27(25)26(23)0.79
Angiographic outcome in PCI patientsn=66n=64
TIMI 2 or 3 pre-intervention (%)43(66)53(82)0.05
TIMI 3 post-intervention (%)59(91)61(94)0.51
Corrected TIMI frame count post23±1418±80.05
  • a CABG=Coronary artery bypass grafting.

  • b PCI=Percutaneous coronary intervention.

7 Enzymatic infarct size and clinical outcome

Results of cardiac enzyme measurements are described in Table 2. Seventy-five percent of patients had a positive troponin T (>0.05ng/ml) on admission or 3h later. In 96 patients (44%), CKmb levels were elevated in the first 48h. The CKmb max was more than three times elevated in 73% of the Early patient group, compared to 54% in the Late patient group (p=0.06). Enzymatic infarct size (LDHQ48) was calculated in 88% (84/96) of patients with a documented elevated CKmb. The mean LDHQ48was 629±503U/l for Early and 432±444U/l for Late treated patients (p=0.02). The mean CKmb max in the 12 patients without known LDHQ48was 91 (54) for the Early group and 46 (35) for the Late group patients (p=0.33).

View this table:
Table 2

Cardiac enzymes

Early Groupn=109Late Groupn=111p
Troponin T pos (%)a80/103(78)72/101(71)0.29
CPKmb pos ≤48h (%)50(46)46(41)
CPKmb max92±9179±680.43
CPKmb max >3 ULN36(73)25(54)0.06
  • a On admission or 3h later.CPKmb=Creatine kinase mb fraction; ULN=Upper limit of normal (>6% of total CPK (if CPK >200U/l (men) and >170U/l (women)); LDHQ48=Enzymatic infarct size (area under the lactate dehydrogenase release over 48h curve).

Clinical outcome is described in Table 3. A periprocedural myocardial infarction was detected in seven patients undergoing PCI (5%) and in 4 patients after CABG (11%). Major bleeding occurred in 24 patients (11%), from which 16 events (67%) were blood transfusions related to CABG. Nine bleeding events were present in the Early group (8%), and 15 in the Late group (14%, p=0.21). At 30 days follow-up, the combined incidence of death, or recurrent infarction was present in 9.2% and 9.0% of patients in the Early and Late group respectively (p=0.97).

View this table:
Table 3

Thirty-day clinical outcome

Early Groupn=109Late Groupn=111p
Death (%)3(3)5(5)0.37
Recurrent myocardial infarction (MI)
PCIarelated (%)4(4)3(2)0.49
CABGbrelated (%)3(3)1(1)0.30
Spontaneous (%)0(0)2(2)0.25
Death or recurrent MI (%)10(9.2)10(9.0)0.97
  • a PCI=Percutaneous coronary intervention.

  • b CABG=Coronary artery bypass grafting.

8 Discussion

This pilot study shows that in patients with non ST elevation acute coronary syndromes, a strategy of delayed angiography after 24–48h of pre-treatment with tirofiban (upstream therapy) is associated with a better angiographic outcome and less initial enzyme release, compared to an immediate invasive strategy of coronary angiography within 12h without pre-treatment with tirofiban. This result was not expected. Very early intervention was expected to reduce infarct size by early opening of an occluded culprit lesion. However, with interim analysis after the inclusion of the first 200 patients, a higher enzymatic infarct size was found for the Early patient group and therefore the study was stopped prematurely after the inclusion of 220 patients.

9 Value of enzymatic infarct size

It is suggested that enzyme rise in patients with acute non-ST elevation coronary syndromes is due to micro-embolization of thrombotic material from a ruptured plaque and that immediate coronary intervention (<12h after admission) of this ‘hot’ plaque may increase distal embolization, resulting in micro-infarctions. By merely taking the incidence of recurrent myocardial infarction as an end point, re-infarctions or infarct-extension may be missed in patients whose enzymes are still raised due to the presenting acute coronary syndrome. The assessment of total CKmb release or enzymatic infarct size over a period of time takes total enzyme release into account and may better quantify the effect of distal micro-embolization, especially in patients whose enzymes have not returned to normal values on the moment of intervention. This study shows that by using total enzyme release over a period of time, it was possible to detect an effect of treatment, which not became evident when the incidence of (recurrent) myocardial infarction was assessed only.

10 Comparison with other studies

The results of our study are consistent with the results of a PRISM-plus sub-study, which showed that the combination of tirofiban plus heparin was associated with lower peak and mean troponin I levels, compared to heparin alone.11To show this effect, only 105 patients were studied, suggesting that by taking total enzyme release as an end point parameter, future studies on new strategies in patients with non-ST elevation acute coronary syndromes, might be performed with a much smaller sample size.

All patients in this study were treated aggressively. Almost all patients underwent angiography <72h and more than 75% of patients were revascularised within 30 days of admission. This revascularisation rate was higher and angiography was performed earlier than the aggressive arms of FRISC 2 and TACTICS TIMI 18.1,2This study is a comparison of two early invasive strategies in which the Early group underwent angiography very early. This aggressive treatment might be the reason for the relatively low rate of death and recurrent MI in bothgroups, despite the high-risk characteristics of the patients.

The recently presented ISAR-COOL study showed a worse outcome in patients cooled-off for a mean of 86h with tirofiban and clopidogrel, compared to very early angiography after 3h of pre-treatment with the same drugs.12This difference was purely driven by a lower incidence of non-fatal MI. However, as stated above, it is very difficult to assess a recurrent myocardial infarction or infarct extension in patients who undergo invasive treatment at a very early stage, when enzymes are still elevated due to the presenting acute coronary syndrome.

11 Clinical outcome

Although angiographic and enzymatic results were better for the delayed angiography and ‘upstream’ therapy strategy, it did not result in a better clinical outcome. Previous studies showed that an invasive strategy is associated with a better clinical outcome, compared to an initial conservative management.1,2The FRISC 2 study even showed a better long-term mortality in these aggressively treated patients.13It is highly ambitious to expect that relatively slight changes in pre-angiography treatment would result in a further improvement in clinical outcome. Apart from the fact that the incidence of recurrent myocardial infarction might be underestimated in the very early treated group, this study lacks power to prove any difference in clinical outcome.

12 Limitations

The main limitation of this study is its design in which both time to angiography and pre-angiography treatment differed between the groups. It remains unknown whether the beneficial effect of the Late strategy is due to the delay itself or due to tirofiban pre-treatment.

Enzymatic infarct size was measured over a period of 48h after hospital admission. The majority of patients in the Late group underwent angiography at a later stage. However, as previously mentioned, assessment of recurrent or periprocedural infarction is especially difficult when enzymes are still elevated due to the presenting acute coronary syndrome during the first 48h after admission. After 48h, enzyme levels have returned to normal and recurrent myocardial infarction beyond this time is more obvious and easier to detect.

13 Conclusion

This pilot study shows that in non ST elevation acute coronary syndrome patients, a strategy of delayed angiography after 24–48h ‘upstream’ therapy with tirofiban is associated with improved angiographic outcomes and a smaller initial enzyme release, compared to a strategy of immediate angiography without 2b/3a pre-treatment. Furthermore it was found that total enzyme release over a given period of time (i.e. enzymatic infarct size) might be of special value in studies concerning non ST elevation ACS, in which patients undergo very early intervention.


View Abstract